Comparative Transcriptome Analyses Reveal Genes Associated with SARS-Cov-2 Infection of Human Lung Epithelial Cells
Total Page:16
File Type:pdf, Size:1020Kb
bioRxiv preprint doi: https://doi.org/10.1101/2020.06.24.169268; this version posted June 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Comparative transcriptome analyses reveal genes associated with SARS-CoV-2 infection of human lung epithelial cells Darshan S. Chandrashekar1, *, Upender Manne1,2,#, Sooryanarayana Varambally1,2,3,#* 1Department of Pathology, University of Alabama at Birmingham, Birmingham, AL 2O’Neal Comprehensive Cancer Center, University of Alabama at Birmingham, Birmingham, AL 3Informatics Institute, University of Alabama at Birmingham, Birmingham, AL *Correspondence to: Sooryanarayana Varambally, Ph.D., Molecular and Cellular Pathology, Department of Pathology, Wallace Tumor Institute, 4th floor, 20B, University of Alabama at Birmingham, Birmingham, AL 35233, USA Phone: (205) 996-1654 Email: [email protected] And Darshan S. Chandrashekar Ph.D., Department of Pathology, University of Alabama at Birmingham, Birmingham, AL Email: [email protected] # Share Senior Authorship (UM Email: [email protected]) Running Title: SARS-CoV-2 gene signature in infected lung epithelial cells Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed. Page | 1 bioRxiv preprint doi: https://doi.org/10.1101/2020.06.24.169268; this version posted June 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Abstract: Understanding the molecular mechanism of SARS-CoV-2 infection (the cause of COVID-19) is a scientific priority for 2020. Various research groups are working toward development of vaccines and drugs, and many have published genomic and transcriptomic data related to this viral infection. The power inherent in publicly available data can be demonstrated via comparative transcriptome analyses. In the current study, we collected high-throughput gene expression data related to human lung epithelial cells infected with SARS-CoV-2 or other respiratory viruses (SARS, H1N1, rhinovirus, avian influenza, and Dhori) and compared the effect of these viruses on the human transcriptome. The analyses identified fifteen genes specifically expressed in cells transfected with SARS-CoV-2; these included CSF2 (colony- stimulating factor 2) and S100A8 and S100A9 (calcium-binding proteins), all of which are involved in lung/respiratory disorders. The analyses showed that genes involved in the Type1 interferon signaling pathway and the apoptosis process are commonly altered by infection of SARS-CoV-2 and influenza viruses. Furthermore, results of protein-protein interaction analyses were consistent with a functional role of CSF2 in COVID-19 disease. In conclusion, our analysis has revealed cellular genes associated with SARS-CoV-2 infection of the human lung epithelium; these are potential therapeutic targets. Introduction: Infection of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is the cause of human coronavirus disease 2019 (COVID-19). The recent pandemic has caused devastation due to rapid spread of this viral infection. As a respiratory illness, the disease is readily transmitted. It also has a long incubation and can be carried asymptomatically, thus spreading through communities (1). The COVID-19 pandemic has affected almost every country regardless of their medical infrastructure and economic status. It has caused a healthcare crisis and created a devastating economic burden, including high unemployment, which has exacerbating the effect of the disease (2). At present, more than 9.4 million people from 213 countries have been infected with the virus [https://www.worldometers.info/coronavirus/]. The rapid spreading of this respiratory infection has forced millions to shelter in their homes and has led to death of more Page | 2 bioRxiv preprint doi: https://doi.org/10.1101/2020.06.24.169268; this version posted June 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. than 480,000 individuals. Additionally, COVID-19 disproportionately affects patients, particularly minorities in the U.S. and those with chronic problems such as hypertension, lung disease, diabetes, and immunocompromised conditions. In the last three decades, the world has witnessed zoonotic transmission of various viruses (from animals to humans) leading to severe respiratory complications. These include H1N1, avian influenza, severe acute respiratory syndrome (SARS), and Middle East respiratory syndrome coronavirus (MERSCoV) (3, 4). Although infections of these viruses is often fatal, their effect is restricted to geographic locations such as Africa, Asia, and South America. As the world awaits a vaccine for SARS-CoV-2, efforts are being made to understand the molecular mechanisms of these infections (5, 6). High-throughput technologies such as RNA sequencing and microarrays are useful in the detection of respiratory virus infections and in understanding their molecular effect on human lung epithelial cells (7). Extensive data on corona virus sequencing has been deposited in public repositories such as NCBI Gene Expression Omnibus (GEO) and EMBL Array Express (8, 9). Meta-analysis and mining of such data can aid in a) understanding the molecular impact of COVID-19, b) elucidating differences and similarities between SARS-CoV-2 and other respiratory virus infections, and c) identification of targets for drug development. In the current study, we performed comparative analysis of publicly available gene expression data related to human lung epithelial cells infected with a respiratory virus. The analyses identified genes specifically expressed by SARS-CoV-2 infection and those that are commonly altered due to infection of coronovirus-2 and/or other respiratory viruses. In particular, expression of CSF2 (colony-stimulating factor 2) appears to be involved in COVID-19 disease. Methods: RNA sequencing data analysis. A search of the NCBI GEO database for microarray or RNA-sequencing data publicly available as of 13th April 2020, found RNA-seq data uploaded by Blanco-Melo et al. (GSE147507) (6). We focused on samples related to normal human bronchial epithelial cells subjected to mock treatment (n=3) or SARS-CoV-2 infection (n=3). Raw sequencing data related to selected samples of GSE147507 as fastq files were downloaded from Sequence Read Archive (SRA) Page | 3 bioRxiv preprint doi: https://doi.org/10.1101/2020.06.24.169268; this version posted June 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. using fastq-dump of sratoolkit v2.9.6 [http://ncbi.github.io/sra-tools/]. First, raw sequencing reads were trimmed to remove adapter sequences and low-quality regions using Trim Galore! (v0.4.1) [http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/]. Trimmed reads were subjected to quality control analysis using FastQC [https://www.bioinformatics.babraham.ac.uk/projects/fastqc/]. Tophat v2.1 was used to map trimmed raw reads to the human reference genome (hg38) (10). All bam files from multiple runs related to the same samples were merged and sorted using SAMtools (Version: 1.3.1) (11). Finally, raw read counts were enumerated for each gene in each sample using HTSeq-count (12). Analysis of differential expression was performed using DESeq2 according to a standard protocol [https://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html] (13). Genes with adj.P-value <0.05 and absolute fold change >= 1.5 were considered as significantly differentially expressed. Gene ontology enrichment analyses of the Differentially Expressed Genes (DEGs) were accomplished by use of the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8 online tool (14). Gene ontology (GO) biological processes with P-values <0.05 and gene counts >2 were considered as significantly enriched. Microarray data collection and analysis. The NCBI GEO database was queried for microarray data related to SARS-CoV infections of human lung epithelial cells. A query (SARS-CoV) AND "Homo sapiens"[porgn] AND ("gse"[Filter] AND ("Expression profiling by array"[Filter])) led to 15 search results. After screening, two studies (GSE47962, GSE17400) were selected. To find microarray data related to SARS-CoV infections of human lung epithelial cells, the GEO database was queried using (((Human lung epithelium) OR (Human bronchial epithelial) AND "Homo sapiens"[porgn] AND ("gse"[Filter] AND "Expression profiling by array"[Filter]))) AND (viral infection AND ("gse"[Filter] AND "Expression profiling by array"[Filter])) AND ("gse"[Filter] AND "Expression profiling by array"[Filter]) AND ("Expression profiling by array"[Filter]). This led to 38 search results, three of which (GSE49840, GSE71766, and GSE48575) were selected for analysis. Table 1 provides sample, platform, and cell line details for all five studies. Since SARS-CoV-2 RNA-seq data included transcriptome profiling after 24 hours of infection, in all microarray studies we only considered samples after 24 hours of viral infection. Page | 4 bioRxiv preprint doi: https://doi.org/10.1101/2020.06.24.169268; this version posted June 30, 2020. The copyright holder for this preprint (which